New York, NY [August 28, 2025]—In a groundbreaking development, researchers at the Icahn School of Medicine at Mount Sinai have unveiled a novel approach to predicting the risk of genetic diseases using artificial intelligence (AI) and routine laboratory tests. This innovative method aims to clarify the implications of rare DNA mutations, a challenge that has long perplexed both doctors and patients. The findings were published in the August 28 online issue of Science.
Traditionally, genetic testing results in a binary diagnosis, often leaving patients with uncertainty about their health outcomes. However, the new approach by Mount Sinai researchers employs machine learning to analyze electronic health records, offering a more nuanced perspective on genetic risk. This advancement promises to enhance the precision of medical predictions, especially for conditions that do not fit neatly into yes/no categories, such as high blood pressure, diabetes, and cancer.
Revolutionizing Genetic Risk Assessment
The Mount Sinai team, led by Ron Do, PhD, and Iain S. Forrest, MD, PhD, has developed AI models that quantify disease risk on a spectrum. These models utilize data from routine lab tests like cholesterol levels and blood counts, which are already part of most medical records. “We wanted to move beyond black-and-white answers that often leave patients and providers uncertain about what a genetic test result actually means,” Dr. Do explained.
By analyzing over one million electronic health records, the researchers constructed AI models for ten common diseases. They applied these models to individuals with known rare genetic variants, generating a “ML penetrance” score between 0 and 1. A higher score suggests a greater likelihood of disease development, while a lower score indicates minimal or no risk.
“By using artificial intelligence and real-world lab data, we can now better estimate how likely disease will develop in an individual with a specific genetic variant,” Dr. Do stated. “It’s a much more nuanced, scalable, and accessible way to support precision medicine.”
Implications for Clinical Practice
The implications of this research are profound. The AI model is not intended to replace clinical judgment but to serve as a guide, particularly when test results are ambiguous. “Doctors could in the future use the ML penetrance score to decide whether patients should receive earlier screenings or take preventive steps,” noted Dr. Forrest. This approach could help avoid unnecessary interventions for low-risk variants while prompting timely action for high-risk cases.
For example, if a patient carries a rare variant associated with Lynch syndrome and receives a high score, it could trigger earlier cancer screenings. Conversely, a low score might prevent undue anxiety and overtreatment. The researchers are now expanding the model to encompass more diseases, a broader range of genetic changes, and more diverse populations.
Looking Ahead: A New Era in Precision Medicine
This development represents a significant step towards integrating AI with routine clinical data to provide personalized, actionable insights. “Our hope is that this becomes a scalable way to support better decisions, clearer communication, and more confidence in what genetic information really means,” Dr. Do expressed optimistically.
The study, titled “Machine learning-based penetrance of genetic variants,” was authored by Iain S. Forrest, Ha My T. Vy, Ghislain Rocheleau, Daniel M. Jordan, Ben O. Petrazzini, Girish N. Nadkarni, Judy H. Cho, Mythily Ganapathi, Kuan-Lin Huang, Wendy K. Chung, and Ron Do. The research was supported by various grants from the National Institutes of Health (NIH), underscoring the collaborative effort to advance precision medicine.
As the team continues to refine their models and track the real-world outcomes of high-risk variants, the potential for AI to transform genetic risk assessment becomes increasingly evident. This pioneering work not only enhances our understanding of genetic diseases but also paves the way for more informed healthcare decisions, ultimately benefiting patients and families navigating the complexities of genetic testing.
